Innate lymphoid cells (ILCs) comprise cytotoxic natural killer (NK) cells and helper ILCs (hILCs). Human hILC development is less characterized as compared with that of NK cells, although all ILCs are developmentally related. It has been reported that the immunosuppressive drugs glucocorticoids (GCs) regulate ILC function, but whether they control ILC differentiation from hematopoietic stem cells (HSCs) is unknown.
This study sought to analyze the effect of GCs on ILC development from HSCs.
This study exploited an in vitro system to generate and expand from peripheral blood HSCs a multipotent CD56+ ILC precursor able to differentiate into NK cells, ILC1s, and ILC3s. We also analyzed ex vivo, at different time points, the peripheral blood of recipients of allogeneic HSC transplantation who were or were not treated with GCs and compared ILC subset reconstitution.
In vitro, GCs favor the generation of NK cells from myeloid precursors, while they strongly impair lymphoid development. In support of these data, recipients of HSC transplantation who had been treated with GCs display a lower number of circulating hILCs, including the ILC precursor (ILCP) previously identified as a systemic substrate for tissue ILC differentiation.
GCs impair the development of the CD117+ ILCP from CD34+ HSCs, while they do not affect the further steps of ILCP differentiation toward NK cells and hILC subsets. This reflects an association of GC treatment with a marked reduction of circulating hILCs in the recipients of HSC transplantation.
BM (Bone marrow), CRTH2 (Chemoattractant receptor-homologous molecule expressed on TH2 cells), Ctrl (Control), GC (Glucocorticoid), G-CSF (Granulocyte-colony stimulating factor), GvHD (Graft versus host disease), HD (Healthy donor), hILC (ILC helper), HSC (Hematopoietic stem cell), HSCT (HSC transplantation), ILC (Innate lymphoid cell), ILCP (ILC precursor), NK (Natural killer), PB (Peripheral blood), pop (Population), ROR (Retinoic acid-receptor–related orphan nuclear receptor), T-BET (T-box expressed in T cells)
While NK cells are circulating in the peripheral blood (PB), hILCs are mainly resident at the mucosal barrier interfaces, where they play a homeostatic and protective role.
NK cells and hILCs are developmentally related and regulate innate immune functions before the development of adaptive immunity.
NK cell development proceeds by steps from CD34+CD45RA+ multipotent HSCs in the bone marrow (BM) (stage 1) to sequential stages that have been characterized in the secondary lymphoid tissues.
These steps are defined by the acquisition of CD56, CD117, and IL-1R1 (stage 2); the loss of CD34 (stage 3);
and the downregulation of CD117 with the acquisition of CD94, which marks commitment to the CD56bright NK cells (stage 4)
that subsequently express CD16 and killer immunoglobulin-like receptors (CD56dim NK) (stage 5); and CD57 (stage 6).
Notably, the myeloid marker CD33 expression can be detected on NK cells from stages 1 to 4, although at lower intensity as compared to myeloid cells.
In this context, it is now recognized that NK cells not only differentiate from lymphoid, but also from myeloid precursors when cultured with NK-supporting cytokines.
This was shown in vitro in cultures from CD34+ HSC isolated from cord blood
and in vivo in humanized mice.
In both settings, CD56+ cells coexpressing myeloid markers such as CD33 and CD14 were shown to undergo maturation toward the conventional NK cell lineage, as evidenced by the progressive downregulation of these markers, and acquisition of cytolytic function and IFN-γ production abilities.
It was shown that stage 2 progenitors in secondary lymphoid tissues are indeed multipotent ILC precursors (ILCPs), which depend on RORγt
and can generate all ILC subsets.
Moreover, while CD117+ ILCs in the PB were previously proposed to represent circulating ILC3s,
they were more recently shown to comprise ILCPs that can give rise to all hILC subsets, as well as to NK cells.
In humanized mice, circulating ILCPs were shown to originate from CD34+ HSCs,
but no further information exists, to date, on their development.
Moreover, hILCs were associated to low/absent graft versus host disease (GvHD), in both human
settings. Therefore, promoting the reconstitution of the hILC compartment in HSCT may be protective and might contribute to a positive transplantation outcome.
but whether it may control their differentiation from HSCs is unknown. In particular, glucocorticoids (GCs) are steroid hormones that, because of their potent anti-inflammatory effects, are given in their synthetic form to patients undergoing allogenic HSCT to prevent or treat immune-related complications, including GvHD.
It was suggested that GCs might enhance the success of HSCT because they increase CXCR4 expression on HSCs and favor their engraftment in the BM.
Moreover, GCs could favor and accelerate NK cell differentiation from myeloid precursors in vitro.
However, it has not been investigated so far whether and how GCs affect the developmental trajectory of CD34+ HSC toward ILCP and from this precursor to all the mature ILC subsets.
We used the PB leukapheresis of granulocyte-colony stimulating factor (G-CSF)–mobilized healthy donors (HDs) as a source of HSCs, to study the effect of GCs on CD34+ HSC differentiation in vitro toward ILCs. Ex vivo, we analyzed, at different time points, the PB of the recipients, comparing ILC subsets reconstitution in children who were or were not treated with GCs. We found that GCs impair the development of the CD117+ ILCP from CD34+ HSCs, while they do not affect the further steps of ILCP differentiation toward NK cells and hILC subsets. This reflects an association of GC treatment with a reduction of circulating hILCs in the recipients of HSCT.
HDs and patients samples
HDs received G-CSF subcutaneously (10-12 μg/kg/day) for 5 days before undergoing leukapheresis. In poor mobilizing donors (circulating CD34+ cell count < 0.04 × 109/L), a single dose of Plerixafor (240 μg/kg) was added to the mobilization regimen. PBMCs were obtained from HD leukapheresis after density gradient centrifugation (Ficoll-Lympholyte, Cedarlane, Burlington, NC). Among HDs, 48% were men (age 40 ± 10 years) and 52% were women (age 38 ± 6 years).
Table IClinical characteristics of the patients analyzed
ALL, Acute lymphoblastic leukemia; AML, acute myeloid leukemia; BU, busulfan; CMV, cytomegalovirus; CY, cyclophosphamide; F, female; FA, Fanconi anemia; FLU, fludarabine; HCV, hepatitis C virus; HHV6, human herpesvirus 6; HLH, hemophagocytic lymphohistiocytosis; M, male; MDS, myelodysplastic syndrome; MEL, melphalan; RCC, refractory cytopenia of childhood; SCID, severe combined immunodeficiency; TBI, total body irradiation; THAL, thalassemia; TREO, treosulfan; TT, thiotepa; WAS, Wiskott-Aldrich syndrome.
For each subject undergoing haploidentical HSCT, sex, diagnosis, conditioning regimen, age at the time of transplantation, grade of acute GvHD (where present), infection (where present), and corticosteroid treatment are reported.
Both HDs and recipients of HSCT gave their informed consent to participate in this study, which was approved by the Bambino Gesù Children’s Hospital (Rome, Italy) ethics committees and was conducted in accordance with the tenets of the Declaration of Helsinki.
Cell isolation, culture, and stimulation
CD34+ HSCs were isolated from mobilized-HD PBMCs by positive selection using CD34 MicroBead Kit UltraPure (Miltenyi Biotech, Bergisch Gladbach, Germany) following the manufacturer’s instructions. Pure CD34+ HSCs were cultured with complete medium (RPMI 1640 containing 10% AB serum, 100 U/ml penicillin, 0.1 mg/mL streptomycin, 2 mmol/L L-glutamine [all from Euroclone, Milan, Italy]) and the cytokines FLT3LG, SCF, IL-7, and IL-15 (20 ng/mL each; Miltenyi Biotech), in the presence of 500 nmol/L dexamethasone (Sigma-Aldrich, St Louis, Mo) or the same volume of vehicle alone (dimethyl sulfoxide). To induce CD56+ ILCP expansion, bulk cultures at day 21 were incubated with IL-7 (20 ng/mL; Miltenyi Biotech) in complete medium for 2 weeks. To induce CD56+ ILCP differentiation, sorted cells were incubated with IL-2 (600 U/mL; Novartis, Basel, Switzerland); IL-7, IL-15, IL-12, IL-1β, IL-23, IL-25, and IL-33 (20 ng/mL each; Miltenyi Biotech); and IL-18 (20 ng/mL; R&D Systems, Minneapolis, Minn).
For NK cell cytokine stimulation, bulk cell cultures at day 28 were incubated at 37°C with IL-12 (20 ng/mL; Miltenyi Biotech) and IL-18 (20 ng/mL; R&D Systems) overnight, Golgi Plug (1:500; BD Biosciences, Franklin Lakes, NJ) and Golgi Stop (1:750; BD Biosciences) were added for the last 4 hours of stimulation. For NK cell degranulation assay, bulk cell cultures at day 28 were incubated at 37°C for 4 hours with P815 cells (a FcγR+ mastocytoma murine cell line) either in the presence or absence of a mix of anti–natural cytotoxicity receptors mAbs (anti-NKp46 BAB281 clone, anti-NKp30 AZ20 clone, anti-NKp44 Z231 clone), with Golgi Plug, Golgi Stop (BD Biosciences), and anti-CD107a-Ab APC (BD Biosciences). For intracellular cytokines detection, cells were stimulated with phorbol 12–myristate 13–acetate (10 ng/mL; Sigma) plus ionomycin (1 mg/mL; Sigma-Aldrich) in the presence of Golgi Plug and Golgi Stop (BD Biosciences) for 3 hours at 37°C.
RNA isolation, library construction, sequencing, and analysis
Differentially expressed genes between GC and Ctrl samples were then identified from the normalized gene-level counts using the following cutoff values: fold-change of 1.5 and adjusted P value of .05. GOrilla (Gene Ontology Enrichment Analysis and Visualization Tool) was utilized to identify GO terms (biological processes) enriched in the differentially expressed genes with respect to the whole list of genes detected (GOrilla – a tool for identifying enriched GO terms [technion.ac.il]). Genomix4Life (Salerno, Italy) processed the RNA samples and performed differential expression bioinformatics analyses.
Antibodies, flow cytometry, and cell sorting
For flow cytometry analysis, cells were first stained with LIVE/DEAD Fixable Blue Dead Cell Stain Kit (Invitrogen, Thermo Fisher Scientific, Waltham, Mass) and surface antibodies in PBS 5% FCS for 20 minutes at 4°C. For intracellular transcription factors and/or cytokines staining, cells were fixed, permeabilized, and stained using FOXP3/Transcription Factor Staining Buffer Kit (Miltenyi Biotech). The following antibodies were used: CD34-APC, CD117-APC, CD33-FITC, CD94-APCVio770, Tbet-APC (Miltenyi Biotech); CD3-APC-AF750, CD19-APC-AF750, CD14-APC-AF750, CD45-KromeOrange, CD34-FITC; CD94-APC, CD56-APC-AF700, CD117-EDC; CD14-ECD, CD56-PE-Cyanine7, CD127-PE, CRTH2 (chemoattractant receptor-homologous molecule expressed on TH2 cells)-FITC, CD158b1/b2,j-PE, CD158e1/e2-PE, CD158a/h-PE, NKp44-PE, IgG1-PE (Beckman Coulter, Indianapolis, Ind); CD56-BV650, NKG2D-BV605, CD16-BV510, GranzymeB-BV421, IgG1-PE-Cyanine7 (BD Biosciences); DNAM-APC, CD117-PerCP/Cyanine5.5, CRTH2-PE, CD94-FITC, Perforin-PE (BioLegend, San Diego, Calif); NKp46-eFluor450, IFN-γ-eFluor450, Eomes-eFluor660, Rorγt-PE, Tbet-PE-Cyanine7, IL-22-PE, Eomes-PE-Cyanine7, Rorγt-APC (eBiosciences, Thermo Fisher Scientific); and Tbet PerCP, IgG2a-APC (R&D Systems).
FlowSOM generated the heatmap for each channel used, and each FlowSOM population (eg, pop 0) identified was overlayed on t-SNE (t-distributed stochastic neighbor embedding) run on concatenated files (shown in Fig 1, B).
Statistical analysis was performed with GraphPad Prism 8 Software (GraphPad Software, San Diego, Calif). Normality was tested with the Shapiro-Wilk test. Unpaired 2-tailed Student t-tests or paired Student t-tests (when data from the same donor were compared) were used to compare 2 groups. One-way ANOVA was used for multigroup comparisons. Multiple t-test was used to compare repeated data from the same donors at different time points. The statistical test used is indicated in each figure legend. Differences were considered significant for P values less than .05.
Data and code availability
The data that support the findings of this study are available from the corresponding authors on request. The RNA sequencing data have been deposited in the GEO repository under accession code GSE186950.
GCs affect CD34+ HSC differentiation toward CD56+ cells in vitro
the most abundant population obtained by culturing BM-mobilized HSC with the cytokine mix was represented by CD33+ cells with myelomonocytic features, expressing or not CD14 (pops 1 and 3) (Fig 1, B-D). Nevertheless, our results show that GCs specifically affect the development of CD56+ cell populations in culture. Indeed, the CD33–CD56high population frequency increased over time in control condition, while it was almost absent in GC cultures (Fig 1, D). On the other hand, the CD33+CD56int population emerged early in culture in GC condition and decreased over time, while its frequency remained below 10% in the control condition (Fig 1, D). No significant differences were found for the other populations, including myelomonocytic pops 1 and 3 (Fig 1, D), and CD33+CD117high pop 0, whose frequency was lower than 1.5% at all time points analyzed (Fig E1, C).
These results suggest that GCs induce transcriptional modifications in CD34+ HSCs cultured with the cytokine mix, specifically affecting the development of CD56+ cells.
GCs induce a preferential differentiation of CD94+ NK cells from myeloid precursors
These results demonstrate that GCs induce a preferential NK cell differentiation from CD33+ myeloid precursors, thus preserving HSC potential to generate functional NK cells in culture.
GCs inhibit the development of a common CD56+ ILCP
which generate not only NK cells, but also other ILC subsets.
Importantly, GC addition to IL-7, IL-15, IL-12, IL-18, IL-1β, and IL-23 did not alter the ILCP differentiation potential (Fig 4, A and D). These results demonstrate that GCs impair the development of a common CD56+ ILCP from CD34+ HSC in vitro, but not its ability to differentiate into NK cells, ILC1s, and ILC3s.
In HSCT, GC treatment is associated to a reduction of circulating hILCs
In this setting, we analyzed a cohort of 44 patients (Table I). Twenty-one of them received GC treatment 3.5 ± 0.6 months post HSCT. GC therapy was used mainly to treat GvHD and in a smaller fraction of patients GCs were used for other clinical complications, including hemolytic anemia, engraftment syndrome, and hemolytic–uremic syndrome (Table E2). Six, 9, 12, and 24 months after transplantation, PBMCs isolated from the patients were analyzed for the presence of ILCs, using the gating strategy shown in Fig 5, A and B. We further confirmed that CD127+CD94−CD117−CRTH2− cells can be identified as ILC1s by demonstrating that they express T-BET but not EOMES (see Fig E4, A, in this article’s Online Repository at www.jacionline.org). The frequency of CD94+ NK cells and of hILCs among lineage-negative PBMCs was not significantly affected by GC treatment (Fig E4, B). Of note, in this transplantation setting, mature NK cells are present in the graft and it has been shown that they persist in the recipient PB in the early phase after the allograft (1 month post-HSCT).
However, at the time points analyzed, circulating NK cells derive exclusively from the differentiation of engrafted HSC, and reach the stage of CD94/NKG2A+ cells in 2 to 3 weeks.
Accordingly, we found that the number of immature CD94+ NK cells in the PB of HSCT recipients did not further increase after 6 months, and no difference was observed between Ctrl and GC-treated children (Fig 5, C), suggesting that GC treatment does not affect NK cell development. Also all circulating CD127+ hILCs in the recipient PB are of donor origin, but their reconstitution is slower compared with that of NK cells.
Our data show, indeed, that reconstitution was still incomplete 6 months post-HSCT and it was fully achieved in patients in the Ctrl group at 12 months, as hILCs numbers remained constant thereafter (Fig 5, C). At the time of complete reconstitution, the number of hILCs in children who had been treated with GCs was significantly lower compared with the number of hILCs in the Ctrl group (Fig 5, C). This defect was mainly due to a GC-dependent decrease in the CD117+ ILCP and CD117− ILC1 subsets (Fig 5, D); however, the proportion of each subset within hILCs was unchanged, with ILC1s representing the most abundant subset (Fig E4, C). Importantly, by further stratifying the data at 12 months post-HSCT for the group of patients treated with GCs (Table E2), a dose-dependent effect of GCs on hILC numbers was observed (see Fig E5 in this article’s Online Repository at www.jacionline.org).
These results are in agreement with our data in vitro, showing that GCs do not alter NK cell differentiation but impair the development of hILCs.
HSCT recipients who experienced infections display higher circulating hILC numbers
Here, we further stratified the cohort of patients that we analyzed 12 months after HSCT according to the emergence of GvHD, to perform a retrospective analysis. All patients who experienced GvHD in the first year had been treated with GCs (Fig 6, A). We confirmed that GvHD is associated to reduced circulating hILC subset (purple vs green violin), although this difference is not statistically significant (Fig 6, B and C). However, patients who had been treated with GCs and did not show GvHD (pink violins) displayed numbers of hILCs comparable to the GvHD group (purple violins) and significantly lower than the control group (green violins) (Fig 6, B and C). These data indicate that GC treatment per se induces a reduction in hILC numbers, which is independent on the development of GvHD.
In our cohort, half of the patients who had infections were later treated with GCs. Virus reactivation/infection and/or bacterial infections had occurred at 1.2 ± 0.36 months post-HSCT in the Ctrl group and 1.8 ± 0.5 months in the GC group. In Fig 6, D, the frequency of previous infections is shown for Ctrl and GC groups. As evidenced by their distribution in Fig 6, B and C, Ctrl subjects’ data are very heterogeneous. By stratifying them on the basis of previous infection criteria, we found that patients who had infections display higher circulating hILC numbers, in particular ILCPs (Fig 6, E and F). However, comparing the GC-treated groups of patients, no difference was found between subjects on the basis of this type of stratification (Fig 6, E and F). Moreover, focusing our analysis only on patients who had infections, we found that the GC-dependent defect in the reconstitution of the hILC compartment is even more evident in this group of subjects, in particular for ILC1s and ILCPs (Fig 6, E and F).
These data suggest that, in recipients of HSCT, the early occurrence of an infection is associated with an enhanced ILCP development, possibly mediated by inflammatory cytokines. In addition, these data collected at 1 year post-HSCT show that GC treatment following infections is able to antagonize this stimulatory pathway in the long term.
NK cells generated in vitro in the absence or in the presence of GCs are quantitatively, phenotypically, and functionally equivalent and can be defined as stage 4 NK cells, resembling CD56bright NK cells according to the linear model of NK cell development.
In recipients of HSCT, NK cell differentiation from HSC requires 2 to 3 weeks to reach a maturation stage analogous to the in vitro stage 4.
The analysis of patients’ PB starting from 6 months post-HSCT supports our findings in vitro. Indeed, the lack of effect of GC treatment on NK cell reconstitution in the long term despite the reduction of circulating ILCP confirms that there are other progenitors from which NK cells can derive. These may include myeloid precursors, as well as CD127−-committed NK cell progenitors identified in BM, blood, and secondary lymphoid tissues .
we show that stage 3 CD56+CD117+CD127+ cells generated in vitro from HSC do not comprise lymphoid NK precursors and mature ILC3s with an overlapping phenotype. These cells (that we define as CD56+ ILCPs) are indeed homogeneous not only from a phenotypic, but also from a functional point of view and have a common developmental potential, as they fail to produce cytokines on stimulation and acquire this ability only when further differentiated. CD56+ ILCPs are therefore analogous to the systemic ILCP identified in the PB, cord blood, and fetal liver,
although with a more restricted potential. It was shown that the expression of CD56 by this progenitor marks the divergence of a shared NK/ILC common developmental pathway from ILC2s.
In addition, NKp46 was identified as a marker that clearly defines the ILC3 potential.
In line with these data, the CD56+ ILCP we have identified in vitro is capable of generating all ILC subsets with the exception of ILC2s. Therefore, our findings reconcile the previous in vitro human ILC developmental model with the more recent identification of systemic ILCPs in the PB. The study of human ILC development is limited by the lack of tools comparable to those available in animal models and by the substantial differences existing with murine ILCs development.
In this context, we report here an in vitro model that recapitulates human ILCs development and allows to obtain IL7-expanded multipotent ILCPs. This may represent a useful feeder-free tool to further dissect the molecular mechanisms that define the distinct pathways by which each human ILC subset is generated, complementary to the in vitro platform recently developed by Hernandez et al.
Moreover, the possibility of a considerable in vitro expansion of these precursors may represent a promising means to improve mucosal repair and immunity in HSCT recipients, through adoptive cell transfer.
but also the analysis showed that complete recovery takes at least 1 year. In agreement with previously published data,
we show that ILC1s are the first hILC to recover because they represent the most abundant subset. A limitation of the present study is the lack of data on ILC reconstitution in tissues, due to the obvious difficulty of having access to mucosal tissue specimens in humans. In this context, our data do not exclude that the different time interval required for NK cell versus hILC reconstitution as assessed in PB may be dependent on the continuous recruitment of circulating hILCs in the tissues. Moreover, the bidirectional trajectory of hILC reconstitution in PB under GC treatment (which seems to increase for 9 months and then come down at 12 and 24 months) may be due to an effect on cell recruitment into tissues. It would be interesting to compare in parallel hILC reconstitution in the PB and tissues to verify these hypotheses.
In a previous study, the reduction of circulating hILCs expressing activation and tissue homing markers has been associated with higher susceptibility to develop GvHD (19). Because GCs are used to treat GvHD, the question of whether the defect in hILCs is associated with the treatment or with GvHD itself remains to be answered. The analysis of the group of patients who did not develop GvHD but were treated with GCs for other clinical complications unequivocally demonstrates that GCs are responsible for the reduction of circulating hILCs.
According to this model, environmental signals ensure the rapid generation of mature ILCs from a systemic ILCP. Because posttransplantation infections have been associated to increased levels of cytokines such as IL-6, IL-8, and IFN-γ,
our results suggest that inflammatory cytokines released on infection may boost both ILCP development from HSC and its further differentiation into ILC subsets. The long-lasting effect of an early infection on the circulating levels and differentiation capacity of ILCPs may be the result of an “epigenetic scar” left by the antimicrobial response.
Indeed, it was demonstrated that microbial products may generate a “trained immunity”
through epigenetic and functional reprogramming of innate progenitors and HSCs in the BM.
Our data show that, in patients treated with GC, however, infection is not associated with hILCs increase in circulation. It is possible that GCs antagonize the modifications induced by the antimicrobial response in HSCs on transplantation, thus permanently affecting their differentiation potential. In this respect, it has been shown that GCs released in the perinatal period affect long-term T-cell–mediated immunity by inducing a stable early “imprinting” at the chromatin level.
a long-lasting defect of hILCs may indeed alter the homeostasis of mucosal tissues.
Innate lymphoid cells: 10 years On.
Cell. 2018; 174: 1054-1066
Helper-like innate lymphoid cells in humans and mice.
Trends Immunol. 2020; 41: 436-452
Human natural killer cell development.
Immunol Rev. 2006; 214: 56-72
The broad spectrum of human natural killer cell diversity.
Immunity. 2017; 47: 820-833
Modeling human natural killer cell development in the era of innate lymphoid cells.
Front Immunol. 2017; 8: 360
A progenitor cell expressing transcription factor RORgammat generates all human innate lymphoid cell subsets.
Immunity. 2016; 44: 1140-1150
Evidence for discrete stages of human natural killer cell differentiation in vivo.
J Exp Med. 2006; 203: 1033-1043
CD57 defines a functionally distinct population of mature NK cells in the human CD56dimCD16+ NK-cell subset.
Blood. 2010; 116: 3865-3874
Cytomegalovirus infection drives adaptive epigenetic diversification of NK cells with altered signaling and effector function.
Immunity. 2015; 42: 443-456
Human NK cell development: one road or many?.
Front Immunol. 2019; 10: 2078
A novel myeloid-like NK cell progenitor in human umbilical cord blood.
Blood. 2003; 101: 3444-3450
Natural killer-cell differentiation by myeloid progenitors.
Blood. 2011; 117: 3548-3558
Methylprednisolone induces preferential and rapid differentiation of CD34+ cord blood precursors toward NK cells.
Int Immunol. 2008; 20: 565-575
Delineation of natural killer cell differentiation from myeloid progenitors in human.
Sci Rep. 2015; 5: 15118
Differentiation of human innate lymphoid cells (ILCs).
Curr Opin Immunol. 2016; 38: 75-85
Human RORgammat(+)CD34(+) cells are lineage-specified progenitors of group 3 RORgammat(+) innate lymphoid cells.
Immunity. 2014; 41: 988-1000
Human innate lymphoid cells.
Blood. 2014; 124: 700-709
Systemic human ILC precursors provide a substrate for tissue ILC differentiation.
Cell. 2017; 168: 1086-1100.e10
Activated innate lymphoid cells are associated with a reduced susceptibility to graft-versus-host disease.
Blood. 2014; 124: 812-821
Evidence of innate lymphoid cell redundancy in humans.
Nat Immunol. 2016; 17: 1291-1299
Human ectoenzyme-expressing ILC3: immunosuppressive innate cells that are depleted in graft-versus-host disease.
Blood Adv. 2019; 3: 3650-3660
Interleukin-22 protects intestinal stem cells from immune-mediated tissue damage and regulates sensitivity to graft versus host disease.
Immunity. 2012; 37: 339-350
Loss of thymic innate lymphoid cells leads to impaired thymopoiesis in experimental graft-versus-host disease.
Blood. 2017; 130: 933-942
Type 2 innate lymphoid cells treat and prevent acute gastrointestinal graft-versus-host disease.
J Clin Invest. 2017; 127: 1813-1825
Helper innate lymphoid cells in allogenic hematopoietic stem cell transplantation and graft versus host disease.
Front Immunol. 2020; 11582098
Neuroendocrine regulation of innate lymphoid cells.
Immunol Rev. 2018; 286: 120-136
New insights into the cell- and tissue-specificity of glucocorticoid actions.
Cell Mol Immunol. 2021; 18: 269-278
Glucocorticoid hormone-induced chromatin remodeling enhances human hematopoietic stem cell homing and engraftment.
Nat Med. 2017; 23: 424-428
Outcome of children with acute leukemia given HLA-haploidentical HSCT after alphabeta T-cell and B-cell depletion.
Blood. 2017; 130: 677-685
Glucocorticoids and the cytokines IL-12, IL-15, and IL-18 present in the tumor microenvironment induce PD-1 expression on human natural killer cells.
J Allergy Clin Immunol. 2021; 147: 349-360
Computational flow cytometry: helping to make sense of high-dimensional immunology data.
Nat Rev Immunol. 2016; 16: 449-462
Differentiation of natural killer (NK) cells from human primitive marrow progenitors in a stroma-based long-term culture system: identification of a CD34+7+ NK progenitor.
Blood. 1994; 83: 2594-2601
Early expression of triggering receptors and regulatory role of 2B4 in human natural killer cell precursors undergoing in vitro differentiation.
Proc Natl Acad Sci U S A. 2002; 99: 4526-4531
The generation of human innate lymphoid cells is influenced by the source of hematopoietic stem cells and by the use of G-CSF.
Eur J Immunol. 2016; 46: 1271-1278
CD56 expression marks human group 2 innate lymphoid cell divergence from a shared NK cell and group 3 innate lymphoid cell developmental pathway.
Immunity. 2018; 49: 464-476.e4
KLRG1 and NKp46 discriminate subpopulations of human CD117(+)CRTH2(-) ILCs biased toward ILC2 or ILC3.
J Exp Med. 2019; 216: 1762-1776
Phenotypic and functional characterization of NK cells in alphabetaT-cell and B-cell depleted haplo-HSCT to cure pediatric patients with acute leukemia.
Cancers. 2020; 12: 2187
NK cells mediate a crucial graft-versus-leukemia effect in haploidentical-HSCT to cure high-risk acute leukemia.
Trends Immunol. 2018; 39: 577-590
Innate lymphoid cell recovery and occurrence of GvHD after hematopoietic stem cell transplantation.
J Leukoc Biol. 2021 April 13; ()
Immune reconstitution post allogeneic transplant and the impact of immune recovery on the risk of infection.
Virulence. 2016; 7: 901-916
Identification of a human natural killer cell lineage-restricted progenitor in fetal and adult tissues.
Immunity. 2015; 43: 394-407
Human NK cells at early stages of differentiation produce CXCL8 and express CD161 molecule that functions as an activating receptor.
Blood. 2012; 119: 3987-3996
Lineage relationships of human interleukin-22-producing CD56+ RORgammat+ innate lymphoid cells and conventional natural killer cells.
Blood. 2013; 121: 2234-2243
Group 3 innate lymphoid cells (ILC3s): origin, differentiation, and plasticity in humans and mice.
Eur J Immunol. 2015; 45: 2171-2182
Cellular pathways in the development of human and murine innate lymphoid cells.
Curr Opin Immunol. 2019; 56: 100-106
An in vitro platform supports generation of human innate lymphoid cells from CD34(+) hematopoietic progenitors that recapitulate ex vivo identity.
Immunity. 2021; 54: 2417-2432.e5
‘ILC-poiesis’: generating tissue ILCs from naive precursors.
Oncotarget. 2017; 8: 81729-81730
ILC-poiesis: ensuring tissue ILC differentiation at the right place and time.
Eur J Immunol. 2019; 49: 11-18
Serum cytokine levels and acute graft-versus-host disease after HLA-identical hematopoietic stem cell transplantation.
Exp Hematol. 2003; 31: 1044-1050
Elevated serum cytokine levels are associated with human herpesvirus 6 reactivation in hematopoietic stem cell transplantation recipients.
J Infect. 2008; 57: 241-248
Cytokine serum levels during post-transplant adverse events in 61 pediatric patients after hematopoietic stem cell transplantation.
BMC Cancer. 2015; 15: 607
Early blood stream infections after BMT are associated with cytokine dysregulation and poor overall survival.
Biol Blood Marrow Transplant. 2018; 24: 1360-1366
Trained innate immunity, epigenetics, and Covid-19.
N Engl J Med. 2020; 383: 1078-1080
Trained immunity: a program of innate immune memory in health and disease.
Science. 2016; 352aaf1098
C/EBPbeta-dependent epigenetic memory induces trained immunity in hematopoietic stem cells.
Cell Stem Cell. 2020; 26: 657-674.e8
Long-term programming of CD8 T cell immunity by perinatal exposure to glucocorticoids.
Cell. 2020; 180: 847-861.e15
Published online: October 21, 2021
Received in revised form:
In Press Journal Pre-Proof
This work was supported by grants awarded by Associazione Italiana per la Ricerca sul Cancro (AIRC)—Special Program Metastatic Disease: The Key Unmet Need in Oncology 5X1000 2018 ID 21147 (L.M. and F.L.), AIRC IG 2017 ID 19920 (L.M.); RC-2020 OPBG (L.M., P.V.). L.Q. has received funding from AIRC and from the European Union’s Horizon 2020 Research and Innovation Programme under the Marie Skłodowska-Curie grant agreement no. 800924. N.T. and F.B. are supported by an AIRC fellowship for Italy.
Disclosure of potential conflict of interest: The authors declare that they have no relevant conflicts of interest.
© 2021 The Authors. Published by Elsevier Inc. on behalf of the American Academy of Allergy, Asthma & Immunology.
For non-commercial purposes:
- Read, print & download
- Redistribute or republish the final article
- Text & data mine
- Translate the article (private use only, not for distribution)
- Reuse portions or extracts from the article in other works
- Sell or re-use for commercial purposes
- Distribute translations or adaptations of the article